Open Access
Add to BibSonomy
Abstract
InGaAs/InP multi-quantum-well nanowires were directly grown on the v-groove-patterned SOI substrate by metal organic chemical vapor deposition. The surface morphology of the nanowires, the thickness of the quantum wells, and the photoluminescence spectra were characterized by scanning electron microscope, transmission electron microscopy, and micro-photoluminescence, respectively. We found in the experiments that the work of removing part of top Si on both sides of the nanowire to further reduce the optical leakage loss could be completed perfectly without complicated processes, such as a lithography process. Numerical simulations showed that the III-V nanowire was able to support an extraordinarily stable optical guided mode with a lower optical leakage loss of 0.21 cm−1 when etching away part of top Si on both sides of the nanowire, and the optical confinement factor of the multi-quantum-well active region was about 8.8%. This approach opens up a way for monolithic photonic integration of III-V compound semiconductors on Si to occur.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
As the size of microelectronic devices is increasingly approaching their physical limits, silicon microelectronic integrated circuits will be subject to a series of restrictions on fundamental physics problems and technology issues in terms of speed, power consumption, integration, reliability, and other aspects [1,2]. While the development of electronic integrated circuits faces enormous challenges, in recent years, optoelectronic integrated circuits which integrate photonic devices with the mature silicon manufacturing technology have been extensively studied and developed, and it is becoming a promising platform for higher speed interconnect and complex higher functionality with low cost [3]. Unfortunately, Si is extremely hard for emitting laser, especially at room temperature, because it is an indirect bandgap semiconductor. In contrast, III-V materials with direct bandgap have high carrier mobility, high light emission efficiency, and have already dominated in optoelectronic devices. Therefore, heterogeneously integrated III-V materials on Si has been extensively studied [4–7].
Directly epitaxial growth of III-V compound semiconductors on Si is an effective solution to achieve silicon photonics and on-chip optical interconnection applications, which is compatible with CMOS process flow, highly scalable, and lower cost [8]. But, there is a large lattice mismatch and a thermal mismatch between III-V materials and Si, leading to a large number of defects during epitaxial growth [9,10]. Moreover, when the polar III-V materials are epitaxially grown on the surface of non-polar Si, it will lead to the appearance of antiphase domain boundaries (APBs) [11,12], thereby degrading the performance of the device. Fortunately, it is possible to obtain good quality III-V compound semiconductors on the Si, by using aspect ratio trapping (ART) technique and v-grooves made up by two Si {111} facets along the <110> direction. During the growth process, group-III and group-V precursors move into the trenches and crystallize on {111} surfaces. APBs can be inhibited by {111} surfaces in the bottom of the v-grooves, and dislocations can be trapped by SiO2 sidewalls [13–16]. However, the photons generated by the III-V materials easily leak into the substrate, leading to no optical mode existing in the III-V core area, because of the refractive indices of III-V materials (e.g. InP, GaAs) smaller than that of Si in the infrared band [17,18]. The reported method for reducing the optical leakage loss is to hollow out the Si under the III-V nanowire, or to completely separate the III-V nanowire from the Si substrate [6,19].
In this paper, we used the ART technique to directly grow InGaAs/InP multi-quantum-well (MQW) nanowires on the v-groove-patterned SOI substrate by metal organic chemical vapor deposition (MOCVD). The surface morphology of the III-V nanowires, the thickness of the quantum wells (QWs), and the photoluminescence (PL) spectra were characterized by scanning electron microscopy (SEM), transmission electron microscopy (TEM), and micro-photoluminescence (μ-PL), respectively. The SOI substrate can prevent the light field from leaking to the substrate, due to the presence of the buried oxide layer between the top Si layer and the substrate. For further reducing the optical leakage loss, the work of removing part of top Si on both sides of the nanowire was completed without complicated processes such as a lithography process. Numerical simulations showed that the III-V nanowire was able to support an extraordinarily stable optical guided mode with a lower optical leakage loss of 0.21 cm−1 when etching away part of top Si on both sides of the nanowire, and the optical confinement factor of the MQW active region was about 8.8%. The wavelength discussed in the simulations was set to be 1550 nm.
2. Structure and epitaxy of III-V MQW nanowires
Figure 1 shows the preparation process of InGaAs/InP MQW nanowires on the SOI substrate, and the details of the process can be found in [18]. A low temperature (LT) GaAs layer was initially deposited at 400 °C as a seeding layer. Then a high temperature (HT) GaAs layer at 630 °C was performed to fill v-grooves, and used as an intermediary layer for following InP growth, owing to the lattice mismatch of GaAs to Si (4%) lower than that of InP to Si (8%). So the next step was a 400 °C LT-InP layer and a 650 °C HT-InP layer growths on the bottom GaAs in trenches, thereby obtaining high-quality InP. The following was the growths of an InGaAs/InP MQW active region and an HT-InP capping layer. The In/Ga gas flow ratio is 6 during the growth of the InGaAs layer.
Fig. 1 The preparation process of InGaAs/InP MQW nanowires on a v-groove-patterned SOI substrate. (a) Preparing 1-μm-thick SiO2 on top Si layer. (b) Preparing 500-nm-wide SiO2 trenches with a 3-μm gap on the wafer. (c) Etching v-grooves in the top Si layer by a KOH solution. (d) Epitaxy growth of the III-V compound semiconductors.
3. Results and discussion
Figure 2(a) is the cross-sectional SEM image of the III-V nanowire grown on the v-groove-patterned SOI substrate. The v-groove in the top Si layer is of 500-nm width and 450-nm depth. The total height from the bottom of the v-groove to the top of the epitaxial material is 1.8 μm. Figure 2(b) is the plan-view SEM image of ~35 μm × 30 μm area. We can observe that the overall morphology of the III-V nanowires is very uniform. The InGaAs/InP QWs are of the “∧” shape, as shown in Fig. 2(c). For each quantum well (QW), the thickness at the top (001) facet is much thicker than that at two {111} surfaces. Diffusion lengths and deposition probabilities of the group-III elements are different on different crystal planes, and the surface energy of the {001} facets is lower than that of {111} facets. Therefore, the growth rate of InGaAs on the {001} facets is higher than that on the {111} facets [20,21]. A more uniform QW thickness can be obtained by carefully adjusting the growth time of each QW and their position within the nanowire [17]. The GaAs and InP inside the v-groove (A region) are used to filter out the defects caused by lattice mismatch and polarity difference. There are many threading dislocations (TDs) at the interface of InP and GaAs due to the lattice mismatch, as shown in Fig. 2(d). Except that one planar defect such as micro twins passes through the MQW active region, the other locations of the MQW active region are almost defect-free. There are some methods for further improvement in the quality of the III-V nanowire to obtain a completely defect-free MQW active region, such as, appropriately reducing the thickness of GaAs and increasing the thickness of InP in the v-groove to more effectively restrict TDs and twins to the A region [14], and optimizing the thickness and growth temperature of the LT-InP layer to improve the crystal quality of the HT-InP (B region).
Fig. 2 (a) The cross-sectional SEM image of a III-V nanowire on the SOI substrate. (b) The plan-view SEM image of the III-V nanowire array. (c) and (d) The cross-sectional TEM images of the InGaAs/InP MQW nanowire. (e) Normalized μ-PL spectra of a III-V MQW nanowire under two different pump power densities at room temperature.
The μ-PL spectra of the III-V MQW nanowire were measured under two different pump power densities, as shown in Fig. 2(e), with normalized PL peak intensity of InP. The pump source was provided by a 633-nm He-Ne laser under continuous wave operation, and focused by a microscope objective lens onto the III-V nanowire surface with a light spot of about 2-μm diameter. The power of the pump source was about 2 mW, measured by a handheld plane detector at the emergent side of the objective lens, which meant the pump power density on the surface of a III-V nanowire is about 15.9 kW/cm2. The full width at half maximum (FWHM) of the μ-PL spectrum is ~320 nm under the pump power density of 15.9 kW/cm2, and there are many small peaks. The FWHM is ~170 nm at the pump power density of 1.59 kW/cm2. The observed μ-PL spectra are mainly influenced by following three factors: (i) the thickness uniformity of the QWs; (ii) the composition uniformity of the QWs; (iii) the density of pump power onto the surface of the III-V nanowire. As shown in Fig. 2(c), the thickness of each QW is gradually increased from the bottom to the top. In addition, the thickness of three QWs on the same horizontal line or the same vertical line is also not uniform. The thinner QW exhibits a larger effective band gap and can emit photons of shorter wavelengths, whereas the thicker QW can emit longer wavelength photons, thus stepped thickness QWs will lead to spectral overlap and multiple peaks [22]. When the InGaAs QWs are grown on the InP layer with a “∧” shape, the top region most probably is Ga rich, and the bottom region could probably be In rich [16,23–25]. Since the components in the ternary QWs are not unique, the μ-PL spectrum broadens and more than one peak appears. When the power of the pump laser is large enough, electrons at lower energy levels in the valence band can be filled into higher energy levels in the conduction band, resulting in larger band gaps. That is, due to the band-filling effect in the semiconductor MQW structure [26,27], the μ-PL spectrum is broadened to shorter wavelengths and the FWHM is increased. Moreover, there are more small peaks in the μ-PL spectrum when the pump power density is increased, because the density of electrons and holes in a QW is a stepped distribution. For traditional InGaAs MQW, the FWHM of the PL spectrum is very narrow (typically 30-50nm). However, for the InGaAs/InP MQW in the nanowire, the FWHM of the PL spectrum is relatively wide, following the above analysis, this is due to the fact that multiple narrow spectra have overlapped each other.
The III-V nanowire on the SOI substrate supports an optical guided mode, according to simulations by finite difference time domain (FDTD) method using software Lumerical Solutions. In FDTD simulation models, the refractive indices of Si, GaAs, InP, and InGaAs were set to 3.47, 3.37, 3.17, and 3.575, respectively [28,29], and the thickness of the MQW active region was set to 50 nm. Figure 3(a) shows the optical transverse mode supported by the III-V nanowire integrated on the SOI substrate, and the simulated effective refractive index (neff) of this mode is ~2.95. Figure 3(b) is the optical field distribution of the mode viewed in the XY plane after 200-μm propagation in the Z direction. If we assume the spatial size of such a mode was equal to the width of the nanowire (500 nm), a 200-μm distance means that the propagation length is 400 times longer than the mode size. The neff is smaller than the refractive index of each material composing it, which means that the optical field is not completely confined in the III-V MQW nanowire. This can also be confirmed from the Fig. 3(b), in which the optical field spreads into the SiO2 sidewalls and the air regions, and some light unavoidably leaks into the top Si layer. But even so, the SiO2 layer under the top Si layer can effectively block the optical field from leaking to the substrate, therefore, an optical guided mode can be formed in the III-V nanowire. There is a problem that cannot be ignored in the current device structure, that is, the leakage of light to the top Si layer is relatively serious, as shown in Fig. 3(c). The total optical power inside the III-V nanowire above the top Si layer is only 6.6% (normalizing the maximum value of total optical power to 1) after the optical mode propagating 200 μm in the Z direction, as shown by the blue dotted line in Fig. 7(a). The optical leakage lossis given by the following equation:
Fig. 3 The FDTD simulation results of a III-V nanowire on the v-groove-patterned SOI substrate. (a) The optical transverse mode (TE00) supported by a III-V nanowire, viewed in the XY plane. (b) The optical mode distribution after a 200-μm propagation, viewed in the XY plane. (c) The optical field distribution along a 200-μm propagation length, viewed in the YZ plane.
Where P1 and P2 are the total optical power in the III-V material above the top Si layer at Z = L1 and Z = L2, respectively. When L1 = 25 μm and L2 = 200 μm, we can obtain the optical leakage loss () is about 136.55cm−1, calculated by Eq. (1). Because the optical leakage loss is one of the most important losses for the III-V nanowire on Si, the simulations in the paper are used to explore the optical leakage loss for the III-V nanowire on the SOI substrate without considering the optical losses caused by other mechanisms. Although the SOI substrate can prevent the optical field from leaking to the substrate, due to the presence of the buried oxide layer between the top Si and the substrate; these simulation results show that there is still a relatively large optical leakage loss in the top Si layer.
In theory, if the top Si on both sides of a III-V nanowire can be etched away, the optical leakage loss will be greatly reduced. In general, processes such as traditional lithography or electron beam lithography (EBL) are required to define the position of the pattern to be etched away. However, in the experiment, we found that if the sidewall of the III-V nanowire is lower than the sidewall of the SiO2, as shown in Fig. 4(a), as long as the etching time is controlled properly, the SiO2 close to both sides of the nanowire can be first etched away by an HF solution, as shown in Fig. 4(b). This is because of the SiO2 at A and B positions in Fig. 4(a) not on the same plane, and the SiO2 at B position will be simultaneously etched laterally and downward. The SiO2 remaining can be used as a mask together with the III-V nanowire for the removal of the top Si. Therefore, the work of removing top Si on both sides of the III-V nanowire can be completed perfectly without complicated processes. It was found that the rate of SiO2 etched by the HF solution was ~10 nm/s. The top Si can be removed by a KOH solution with a mass fraction of 45%, as shown in Figs. 4(c) and 4(d). To verify whether this method can effectively reduce the optical leakage loss, a III-V nanowire model on the SOI substrate was numerically simulated by the FDTD program, after etching away part of top Si on both sides of the nanowire. The size of the III-V nanowire used in the optical simulation in Fig. 5 is substantially the same as that of the III-V nanowire in Fig. 3. Figure 5(a) is the optical transverse mode supported by the III-V nanowire on the SOI substrate viewed in the XY plane, after etching away part of top Si on both sides of the nanowire. After the light propagates 200 μm in the Z direction, the optical field distribution of the mode viewed in the XY plane is shown in Fig. 5(b), and the optical field distribution viewed in the YZ plane along the propagation direction is shown in Fig. 5(c). The maximum optical field intensity on each XY plane is almost 1 (the maximum value is normalized to 1) in the process of propagating 200 μm in the Z direction. The red dotted line in Fig. 7(a) is the total optical power change inside the III-V nanowire in Fig. 5 along the propagation direction, and the leakage loss () is about 0.21 cm−1 calculated by Eq. (1). Compared to the III-V nanowire without wet etching top Si, the optical leakage loss of the III-V nanowire with wet etching top Si is greatly reduced, which means that the III-V nanowire can support an extraordinarily stable optical guided mode after wet etching top Si on both sides of the III-V nanowire.
Fig. 4 (a) For this III-V nanowire structure, the SiO2 close to both sides of the nanowire can be first etched away. (b) The SEM image of a III-V nanowire on the SOI substrate after etching SiO2 for 40 seconds with an HF solution. (c) and (d) are the SEM images of III-V nanowires on the SOI substrate after etching SiO2 for 60 seconds by an HF solution and etching the top Si for 26 minutes by a KOH solution with a mass fraction of 45%.
Fig. 5 The FDTD simulation results of a III-V nanowire on the SOI substrate after etching away part of top Si on both sides of the nanowire. (a) The optical transverse mode (TE00) supported by the III-V nanowire, viewed in the XY plane. (b) The optical mode distribution after a 200-μm propagation, viewed in the XY plane. (c) The optical field distribution along a 200-μm propagation length, viewed in the YZ plane.
In order to continue to explore the optical leakage loss of the III-V nanowire, the III-V nanowire in Fig. 3 was completely exposed to air for numerical simulation. Figure 6(a) is the optical transverse mode supported by the III-V nanowire in the air, viewed in the XY plane. Figure 6(b) is the optical field distribution of the mode viewed in the XY plane after 200-μm propagation in the Z direction. Figure 6(c) is the optical field distribution along a 200-μm propagation length inside the III-V nanowire, viewed in the YZ plane. The optical leakage loss of the III-V nanowire in the air is smaller than that of the III-V nanowire on the SOI substrate, since the refractive index of the air is less than that of the optical mode. The green dotted line in Fig. 7(a) is the total optical power change in a III-V nanowire completely exposed to air along the propagation direction, and the optical leakage loss () is less than 0.012 cm−1. The optical confinement factors in Fig. 7(b) are obtained by comparing the optical energy in the MQW active region with the total energy in the simulated region. The optical confinement factor in the MQW active region is about 7%, when the top Si on both sides of the III-V nanowire is not etched away, as shown by the blue dotted line in Fig. 7(b). After wet etching the top Si on both sides of the III-V nanowire, the optical confinement factor in the MQW active region is about 8.8%, which is almost the same as that of the MQW active region in the III-V nanowire exposed to air. However, the III-V nanowire completely exposed in the air is not suitable for optoelectronic integration. In contrast, the III-V nanowire in Fig. 4(d) on the SOI substrate is more suitable for optoelectronic integration, under the condition of etching away part of top Si on both sides of the nanowire. The simulation and analysis results in the article are discussed for the case of the TE00 mode. While for the TM00 mode and other high order modes inside the III-V nanowire, the optical leakage losses are also greatly reduced after etching away part of top Si layer, similar to the case of the TE00 mode.
Fig. 6 The FDTD simulation results of a III-V nanowire fully exposed to the air. (a) The optical transverse mode (TE00) supported by the III-V nanowire, viewed in the XY plane. (b) The optical mode distribution after a 200-μm propagation, viewed in the XY plane. (c) The optical field distribution along a 200-μm propagation length inside the III-V nanowire, viewed in the YZ plane.
Fig. 7 (a) The attenuation of the total optical power inside the III-V nanowire above the top Si layer in the three structures after a 200-μm propagation, were obtained by simulation respectively. (b) The optical confinement factors in the three structures were obtained by simulation respectively.
4. Conclusions
InGaAs/InP MQW nanowires were directly grown on the v-groove-patterned SOI substrate by MOCVD using ART technique. SEM was performed to analyze the nanowire’s surface morphology. The thickness of the quantum wells was characterized by TEM. The observed μ-PL spectra are mainly influenced by the thickness uniformity of the QWs, the composition uniformity of the QWs, and the pump power density irradiated onto the surface of the III-V nanowire. The SOI substrate can prevent the light field from leaking to the substrate, due to the presence of the buried oxide layer between the top Si layer and the substrate. For further reducing the optical leakage loss, the work of removing part of top Si on both sides of the nanowire was completed without complicated processes such as a lithography process. The numerical simulation results further confirm that, under the condition of etching away part of top Si on both sides of the nanowire, the InGaAs/InP MQW nanowire on the SOI substrate can support an extraordinarily stable optical guided mode with a lower optical leakage loss and a larger optical confinement factor, and it is more suitable for optoelectronic integration. These research findings open up a possible way for the monolithic integration of Si and III-V compound semiconductors, contributing to the development of nanoscale in photonics and optoelectronic devices.
Funding
Frontier Science Research Project of CAS (Grant No. QYZDY-SSW-JSC021); National Natural Science Foundation of China (Grant No. 61504137); the National Key Technology R&D Program (Grant No. 2018YFA0209001).
Acknowledgments
SOI wafers were obtained from Professor Guangzhao Ran’s group at Peking University.
References
1. S. Chen, W. Li, J. Wu, Q. Jiang, M. Tang, S. Shutts, S. N. Elliott, A. Sobiesierski, A. J. Seeds, I. Ross, P. M. Smowton, and H. Liu, “Electrically pumped continuous-wave III-V quantum dot lasers on silicon,” Nat. Photonics 10(5), 307–311 (2016). [CrossRef]
2. H. Duprez, A. Descos, C. Jany, C. Seassal, and B. Ben Bakir, “Hybrid III-V on silicon laterally coupled distributed feedback laser operating in the o–band,” IEEE Photonics Technol. Lett. 28(18), 1920–1923 (2016). [CrossRef]
3. D. Liang and J. E. Bowers, “Recent progress in lasers on silicon,” Nat. Photonics 4(8), 511–517 (2010). [CrossRef]
4. Y. Sun, K. Zhou, Q. Sun, J. Liu, M. Feng, Z. Li, Y. Zhou, L. Zhang, D. Li, S. Zhang, M. Ikeda, S. Liu, and H. Yang, “Room-temperature continuous-wave electrically injected InGaN-based laser directly grown on Si,” Nat. Photonics 10(9), 595–599 (2016). [CrossRef]
5. L. Yuan, L. Tao, H. Yu, W. Chen, D. Lu, Y. Li, G. Ran, and J. Pan, “Hybrid InGaAsP-Si evanescent laser by selective-area metal-bonding method,” IEEE Photonics Technol. Lett. 25(12), 1180–1183 (2013). [CrossRef]
6. Z. Wang, B. Tian, M. Pantouvaki, W. Guo, P. Absil, J. Van Campenhout, C. Merckling, and D. Van Thourhout, “Room-temperature InP distributed feedback laser array directly grown on silicon,” Nat. Photonics 9(12), 837–842 (2015). [CrossRef]
7. M. Li, L. Zhang, H. Yu, L. Yuan, Q. Kan, W. Chen, Y. Ding, S. Li, J. Mi, G. Ran, and J. Pan, “A Hybrid Single-Mode Laser Based on Slotted Silicon Waveguides,” IEEE Photonics Technol. Lett. 28(9), 1 (2016). [CrossRef]
8. B. Kunert, W. Guo, Y. Mols, B. Tian, Z. Wang, Y. Shi, D. Van Thourhout, M. Pantouvaki, J. Van Campenhout, R. Langer, and K. Barla, “III/V nano ridge structures for optical applications on patterned 300 mm silicon substrate,” Appl. Phys. Lett. 109(9), 091101 (2016). [CrossRef]
9. Y. Bogumilowicz, J. M. Hartmann, N. Rochat, A. Salaun, M. Martin, F. Bassani, T. Baron, S. David, X. Y. Bao, and E. Sanchez, “Threading dislocations in GaAs epitaxial layers on various thickness Ge buffers on 300 mm Si substrates,” J. Cryst. Growth 453, 180–187 (2016). [CrossRef]
10. D. Andrijasevic, M. Austerer, A. M. Andrews, P. Klang, W. Schrenk, and G. Strasser, “Hybrid integration of GaAs quantum cascade lasers with Si substrates by thermocompression bonding,” Appl. Phys. Lett. 92(5), 51117 (2008). [CrossRef]
11. C. Prohl, H. Döscher, P. Kleinschmidt, T. Hannappel, and A. Lenz, “Cross-sectional scanning tunneling microscopy of antiphase boundaries in epitaxially grown GaP layers on Si(001),” J. Vac. Sci. Technol. A 34(3), 031102 (2016). [CrossRef]
12. R. Alcotte, M. Martin, J. Moeyaert, R. Cipro, S. David, F. Bassani, F. Ducroquet, Y. Bogumilowicz, E. Sanchez, Z. Ye, X. Y. Bao, J. B. Pin, and T. Baron, “Epitaxial growth of antiphase boundary free GaAs layer on 300 mm Si(001) substrate by metalorganic chemical vapour deposition with high mobility,” APL Mater. 4(4), 046101 (2016). [CrossRef]
13. J. Z. Li, J. Bai, J. S. Park, B. Adekore, K. Fox, M. Carroll, A. Lochtefeld, and Z. Shellenbarger, “Defect reduction of GaAs epitaxy on Si (001) using selective aspect ratio trapping,” Appl. Phys. Lett. 91(2), 021114 (2007). [CrossRef]
14. M. Paladugu, C. Merckling, R. Loo, O. Richard, H. Bender, J. Dekoster, W. Vandervorst, M. Caymax, and M. Heyns, “Site selective integration of III-V materials on Si for nanoscale logic and photonic devices,” Cryst. Growth Des. 12(10), 4696–4702 (2012). [CrossRef]
15. E. A. Fitzgerald and N. Chand, “Epitaxial necking in GaAs grown on pre-patterned Si substrates,” J. Electron. Mater. 20(7), 839–853 (1991). [CrossRef]
16. S. Li, X. Zhou, M. Li, X. Kong, J. Mi, M. Wang, W. Wang, and J. Pan, “Ridge InGaAs/InP multi-quantum-well selective growth in nanoscale trenches on Si (001) substrate,” Appl. Phys. Lett. 108(2), 021902 (2016). [CrossRef]
17. L. Megalini, B. Bonef, B. C. Cabinian, H. Zhao, A. Taylor, J. S. Speck, J. E. Bowers, and J. Klamkin, “1550-nm InGaAsP multi-quantum-well structures selectively grown on v-groove-patterned SOI substrates,” Appl. Phys. Lett. 111(3), 032105 (2017). [CrossRef]
18. Z. Li, M. Wang, X. Fang, Y. Li, X. Zhou, H. Yu, P. Wang, W. Wang, and J. Pan, “Monolithic integration of InGaAs/InP multiple quantum wells on SOI substrates for photonic devices,” J. Appl. Phys. 123(5), 053102 (2018). [CrossRef]
19. Y. Han, Q. Li, S. Zhu, K. W. Ng, and K. M. Lau, “Continuous-wave lasing from InP/InGaAs nanoridges at telecommunication wavelengths,” Appl. Phys. Lett. 111(21), 212101 (2017). [CrossRef]
20. T. Schrimpf, P. Bönsch, D. Wüllner, H.-H. Wehmann, A. Schlachetzki, F. Bertram, T. Riemann, and J. Christen, “InGaAs quantum wires and wells on V-grooved InP substrates,” J. Appl. Phys. 86(9), 5207–5214 (1999). [CrossRef]
21. S. Li, X. Zhou, X. Kong, M. Li, J. Mi, J. Bian, W. Wang, and J. Pan, “Evaluation of growth mode and optimization of growth parameters for GaAs epitaxy in V-shaped trenches on Si,” J. Cryst. Growth 426, 147–152 (2015). [CrossRef]
22. X. Chen, B. Zhao, Z. Ren, J. Tong, X. Wang, X. Zhuo, J. Zhang, D. Li, H. Yi, and S. Li, “Advantages of InGaN/GaN multiple quantum well solar cells with stepped-thickness quantum wells,” Chin. Phys. B 22(7), 078402 (2013). [CrossRef]
23. Y. Han, Q. Li, K. W. Ng, S. Zhu, and K. M. Lau, “InGaAs/InP quantum wires grown on silicon with adjustable emission wavelength at telecom bands,” Nanotechnology 29(22), 225601 (2018). [CrossRef] [PubMed]
24. S. Jiang, C. Merckling, W. Guo, N. Waldron, M. Caymax, W. Vandervorst, M. Seefeldt, and M. Heyns, “Evolution of (001) and (111) facets for selective epitaxial growth inside submicron trenches,” J. Appl. Phys. 115(2), 23517 (2014). [CrossRef]
25. G. Biasiol, A. Gustafsson, K. Leifer, and E. Kapon, “Mechanisms of Self-Ordering in Nonplanar Epitaxy of Semiconductor Nanostructures,” Phys. Rev. B Condens. Matter Mater. Phys. 65(20), 205306 (2002). [CrossRef]
26. Q. Huang, Y. Wu, K. Ma, J. Zhang, W. Xie, X. Fu, Y. Shi, K. Chen, J.-J. He, D. Van Thourhout, G. Roelkens, L. Liu, and S. He, “Low driving voltage band-filling-based III-V-on-silicon electroabsorption modulator,” Appl. Phys. Lett. 108(14), 141104 (2016). [CrossRef]
27. L. Wang, C. Lu, J. Lu, L. Liu, N. Liu, Y. Chen, Y. Zhang, E. Gu, and X. Hu, “Influence of carrier screening and band filling effects on efficiency droop of InGaN light emitting diodes,” Opt. Express 19(15), 14182–14187 (2011). [CrossRef] [PubMed]
28. E. Palik, Handbook of Optical Constants of Solids (Academic, 1997).
29. H. Dinges, H. Burkhard, R. Lösch, H. Nickel, and W. Schlapp, “Refractive indices of InAlAs and InGaAs/InP from 250 to 1900 nm determined by spectroscopic ellipsometry,” Appl. Surf. Sci. 54, 477–481 (1992). [CrossRef]
